Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Apple Releases Open Weights Video Model

starflow-v.github.io

Post reply on HN
Print view View original

31–40 of 178 posts

Re: Apple Releases Open Weights Video Model

#31

Post by dzhiurgis »

gostsamo wrote: post #27
phyzix5761 wrote: post #9

Earlier quoted context omitted.

Can you share some ways AI has changed your life?

Not the gp, but currently reading a web novel with a card game where the author didn't include alt text in the card images. I contacted them about it and they started, but in the meantime ai was a big help. all kinds of other images on the internet as well when they are significant to understanding the surrounding text. better search experience when Google, DDG, and the like make finding answers difficult. I might us…

As a (web app) developer I never quite sure what to put in alt. Figured you might have some advice here?

Re: Apple Releases Open Weights Video Model

#32

Post by dragonwriter »

mdrzn wrote: post #23

"VAE: WAN2.2-VAE" so it's just a Wan2.2 edit, compressed to 7B.

> "VAE: WAN2.2-VAE" so it's just a Wan2.2 edit

No, using the WAN 2.2 VAE does not mean it is a WAN 2.2 edit.

> compressed to 7B.

No, if it was an edit of the WAN model that uses the 2.2 VAE, it would be expanded to 7B, not compressed (the 14B models of WAN 2.2 use the WAN 2.1 VAE, the WAN 2.2 VAE is used by the 5B WAN 2.2 model.)

Re: Apple Releases Open Weights Video Model

#33

Post by gostsamo »

dzhiurgis wrote: post #31
gostsamo wrote: post #27

Earlier quoted context omitted.

Not the gp, but currently reading a web novel with a card game where the author didn't include alt text in the card images. I contacted them about it and they started, but in the meantime ai was a big help. all kinds of other images on the internet as well when they are significant to understanding the surrounding text. better search experience when Google, DDG, and the like make finding answers difficult. I might us…

As a (web app) developer I never quite sure what to put in alt. Figured you might have some advice here?

The question to ask is, what a sighted person learns after looking at the image? The answer is the alt text. E.g if the image is a floppy, maybe you communicate that this is the save button. If it shows a cat sleeping on the windowsill, the alt text is yep: "my cat looking cute while sleeping on the windowsill".

Re: Apple Releases Open Weights Video Model

#35

Post by dragonwriter »

Mashimo wrote: post #18
M4v3R wrote: post #11

Earlier quoted context omitted.

These are ~2 years behind state of the art from the looks of it. Still cool that they're releasing anything that's open for researchers to play with, but it's nothing groundbreaking.

But 7b is rather small no? Are other open weight video models also this small? Can this run on a single consumer card?

> But 7b is rather small no?

Sure, its smallish.

> Are other open weight video models also this small?

Apples models are weights-available not open weights, and yes, WAN 2.1, as well as the 14B models, also has 1.3B models; WAN 2.2, as well as the 14B models, also has a 5B model (the WAN 2.2 VAE used by Starflow-V is specifically the one used with the 5B model.) and because the WAN models are largely actually open weights models (Apache 2.0 licensed) there are lots of downstream open-licensed derivatives.

> Can this run on a single consumer card?

Modern model runtimes like ComfyUI can run models that do not fit in VRAM on a single consumer card by swapping model layers between RAM and VRAM as needed; models bigger than this can run on single consumer cards.

Re: Apple Releases Open Weights Video Model

#36

Post by justinclift »

postalcoder wrote: post #5
nothrowaways wrote: post #4

Where do they get the video training data?

From the paper: > Datasets. We construct a diverse and high-quality collection of video datasets to train STARFlow-V. Specifically, we leverage the high-quality subset of Panda (Chen et al., 2024b) mixed with an in-house stock video dataset, with a total number of 70M text-video pairs.

> in-house stock video dataset

Wonder if "iCloud backups" would be counted as "stock video" there? ;)

Re: Apple Releases Open Weights Video Model

#37

Post by fragmede »

justinclift wrote: post #36
postalcoder wrote: post #5

Earlier quoted context omitted.

From the paper: > Datasets. We construct a diverse and high-quality collection of video datasets to train STARFlow-V. Specifically, we leverage the high-quality subset of Panda (Chen et al., 2024b) mixed with an in-house stock video dataset, with a total number of 70M text-video pairs.

> in-house stock video dataset Wonder if "iCloud backups" would be counted as "stock video" there? ;)

Turn on advanced data protection so they don't train on yours.

Re: Apple Releases Open Weights Video Model

#38

Post by anon7000 »

justinclift wrote: post #36
postalcoder wrote: post #5

Earlier quoted context omitted.

From the paper: > Datasets. We construct a diverse and high-quality collection of video datasets to train STARFlow-V. Specifically, we leverage the high-quality subset of Panda (Chen et al., 2024b) mixed with an in-house stock video dataset, with a total number of 70M text-video pairs.

> in-house stock video dataset Wonder if "iCloud backups" would be counted as "stock video" there? ;)

I have to delete as many videos as humanly possible before backing up to avoid blowing through my iCloud storage quota so I guess I’m safe

Re: Apple Releases Open Weights Video Model

#40

Post by swores »

tippa123 wrote: post #19
densh wrote: post #7

Earlier quoted context omitted.

> As a blind person, AI has changed my life. Something one doesn't see in news headlines. Happy to see this comment.

+1 and I would be curious to read and learn more about it.

A blind comedian / TV personality in the UK has just done a TV show on this subject - I haven't seen it, but here's a recent article about it: https://www.theguardian.com/tv-and-radio/2025/nov/23/chris-m...
Post reply on HN